ArticlePDF Available

Why Didn't She Just Report Him? The Psychological and Legal Implications of Women's Responses to Sexual Harassment

DOI:10.1111/j.1540-4560.1995.tb01312.x
Authors:
Louise Fitzgerald at University of Illinois, Urbana-Champaign
Louise Fitzgerald
Suzanne Swan
Suzanne Swan
Suzanne Swan
  • This person is not on ResearchGate, or hasn't claimed this research yet.
Karla Fischer at University of Illinois, Urbana-Champaign
Karla Fischer
Download full-text PDF
Read full-text
Download citation

Abstract

The last few years have seen an increasing awareness of sexual harassment as an important social problem with serious implications for individuals and organizations alike, leading to increased attempts to understand how victims respond to this stressful and sometimes traumatic experience. The present article reviews the behavioral science research on responses to sexual harassment, including their links to outcomes and consequences. We then present an alternative to the frequently invoked assertiveness paradigm, derived from the cognitive-behavioral stress and coping framework. We examine our paradigm in the context of legal proceedings that, in effect, hold the victim responsible for responding appropriately; explore the more general implications of placing the burden of noncom-sent on the victim; and conclude with a discussion of this research for an emerging legal theory of sexual harassment.
Content uploaded by Louise Fitzgerald
Author content
All content in this area was uploaded by Louise Fitzgerald on Feb 27, 2019
Content may be subject to copyright.
Joumal
4
Social
Issues,
Vol.
51, No, 1,
1995,
pp. 117-138
Why Didn't She Just Report Him?
The Psychological and Legal ImfiUcations
of Women's Responses to Sexual Harassment
Louise F. Fitzgerald and Suzanne Swan
University of Illinois at Urbana—Champaign
Karla Fbcher
Duke University
The
last few years have seen an increasing awareness <^ sexual harassment as an
important
social problem with serious implications for individuals and organiza-
tions alike, leading to increased attempts to understand how victims respond to
this stressful
and sometimes traumatic experience. The present article reviews the
behavioral
science research on responses to sexual harassment, including their
litiks to outcomes and consequences. We then present an alternative to the fre-
quently invoked assertiveness paradigm, derived from the cognitive-behavioral
stress
and coping framework. We examine our paradigm in the context of legal
proceedings that, in effect, hold the victim responsible for responding appro-
priately; explore the more general implications of placing the burden cfnoncon-
sent on the victim; and conclude with a discussion of this research for an
emerging legal theory of sexual harassment.
If you've been sexuaUy harassed, you ought
to
complain!
... I
mean, where's
the
gumption?—Senator Dennis DeConcini
On the first Sunday of October 1991, National Public Radio reported that
Anita Hill, a professor of law at the University of Oklahoma, claimed to have
been sexually harassed by Clarence Thomas, a hitherto little-known federal
judge then literally hours away bom confirmation as an Associate Justice of the
Supreme Court. The controversy stimulated by those charges and tiie manner in
CoiTespondeace regarding tbis article should be addressed to Louise F. Fitzgerald, Department
of Psychology, Univeisity <rf ffiinois, 603 E. Daniel Street, Oiampaign, IL 61820.
117
Q022-4S37r9Sf03Q0.OI 17SCI3.D0 C VS^i Tbe Soany for Ihc rsyctaokl(icil Study of Social Issues
118 Rbsgenld ct fll.
which Congress dealt with them created a furious national debate, elevating
sexual harassment from its previous status as smutty joke to its present standing
as a major social issue.
The focus of the hearings was credibility, principally tiiat of Professor Hill.
Newspapers, radio, and media of every stripe conducted polls both scientific and
otherwise, each posing the question, "Do you believe Anita Hill?" Unaramously,
die majority of respondents replied that they did not. Despite extMisive evidence
that Hill was a woman of character and integrity, credible witnesses whsitestified
she had confided in them at the time, and the lack of any remotely convincing
native for her to lie, the msycnity of the American public confided to the
pollsters that tlKy simply didn't believe her. A major factor cited by the public
was Hill's "failure" to file any sort of charge against Thomas at the time of the
alleged incidents. If it had really happened, they asked, why didn't she com-
plain? Why didn't she just report him? Many men stoutly asserted that no
self-
respecting woman would have put up with such behavior, and women £^reed that
they personally would never have done so. The climate of disbelief attending this
refusal to respond in a manna- the public found acceptable played no small part
in the outcome of the Senate drama, a drama that continues to unfold in d»e
courtnx>ms of America, with less publicity but similar results.
Our article begins by reviewing the research on responses to harassment,
including their links to outcomes and consequences. We tiien critique the asser-
tion that confrontation and reporting are invariably both appropriate and effec-
tive, proposing that victim responses are complex and multidimensional, and
most usefiilly conceptualized within the cognitive-behavioral frameworic (La-
zarus & Folkman, 1984) that is tbe dominant psychological paradigm for under-
standing stressful life events. We locate this research in the conteiS of legal
proceedings that, in practice if not theory, hold the victim responsible for re-
sponding "ajqw^opdately," explore the more general inqilications of placing the
burden of nonconsent on the victim, and conclude with a discussion of tiie
implications of this research for an emerging legal thecMy of sexual harassment.
VktHn Hespomes to Sexual Hm^ssment: A Brief Review
Classification of Responses
A number of schemes for classifying responses to harassment tove been
proposed (Gruber, 1989; Maypole, 19^; Trapstra & Baker, 1989), sji^nas that
focus mainly on the degree to which the victim responds asseitiv^ to
Hae
harasso:. Gniber's system is a case in point using the Maypole fram^Kork as a
point of dqiaitiire, Gruber ordered responses along a continuum &ofi| lhe least
confiontatiooal to the most. Teipstra and Baker (1989) proposed a
siis&pa:
fcarae-
wmk and Gi^ek aid Koss (1993) provided a two-<fin}ensional system ^ifiaed by
degree of assertiveness, and whether tiie response involved the help 1^ others.
SespoBses to Harassment 119
Although useM as a starting point for theory, such frameworks suffer from
a major shortctnning: diey are not derived from the reactions of actual victims.
With the exception of Maypole's early system, they are based either on rational
derivation (Gruber, 1989; Gutek & Koss, 1993) or, problematically, on the
written responses of research participants to brief descriptions of hypotheti-
cal situations. Given Aat actual victims have been shown to behave quite differ-
ently than research participants or the general public say they would behave
(FitzgCTald, 1993; Gutek & Koss, 1993), the generalizability of such systems is
open to question. Similarly problematic is the focus in this literature on action-
oriented, problem-solving responses to the neglect of more psychological reac-
tions (e.g., cognitive distancing, reattribution), reflecting a widespread assump-
tion that assertive responding is always both appropriate and effective.
Addressing these issues, Fitzgerald (1990) proposed a framework derived
by coding responses provided by actual victims surveyed in a large sexual harass-
Tient prevalence study (Fitzgerald et al., 1988). The system consists of 10
strategies, classified as either internally focused (endurance, denial, detachment,
reattribution, and illusory control) or externally focused (avoidance, appease-
ment, assertion, seeking institutional or organizational
relief,
and seeking social
support). Internal strategies are characterized by attempts to manage the cogni-
tions and emotions associated with the event (e.g., "I just tried to forget about
it"; "I told myself he didn't mean to upset me") whereas extemally focused
strategies arc problem solving in nature (e.g., "I told him to leave me alone"; "I
reported him to the supervisor").
TTje system differs from altemative frameworks principally by the inclusion
of the cognitive strategies that actual victims frequently employ; some readers
will recognize the obvious parallel to the more general strategies (i.e., emotion
focused and {woblem focused) identified by Lazams and Folkman (1984) and
subsequcndy elabcvated by Folkman and Lazams (1989). Confirmatory factor
analysis of an inventory designed to assess these strategies confirms the structure
of the two genoal categories, which are negatively correlated but by no means
mutuidly exclusive. Thus, unlike unidimensional continuum-of-assertiveness
frameworks, tiiis model reflects tiie multidimensional nature of these complex
behavioral phenomraa. In tiie following pages, we employ this framework to
organize our review of the various response strategies and their prevalence,
correlates, and outcomes. Because of the discrepancy between the responses
of analogue partidpaots and those of actual victims, only the latter are re-
viewed here.
Differential Prevalence of Response Strategies
Imemalfy focused responses. A common response, particularly to less se-
vere situations, is sinqtly to ignore tiie harassioent and do nothing (endunmce),
or to pietend tiiat the situation is not happening or has no effect [denial;
120 ftegeraU tt al.
Fit^eiald et al., 1988; Gniber & Bjcm, 1982; Omck, 198S; Goteic & Koss,
1993;
Loy & StsewMt. 1984; U.S. Merit Systems Protection Board (USMSPB),
1981,
1988}. For exanLq>K a British survey found &mt 26% of harassed iMoikers
"did nothing" whatsoever in response to l»ing harassed ^hilUps, Stockdale, &
Joeman, 1989). It is difficult to ascertain exactly what "doing notbi^" means in
this context; for example, it may in^y a deliberate decision to ignore the
situation, to pretend it is not Imppeniag, or that one does not care. Vfe label such
respcmses denial, following Breznitz (1983), who noted tiiat denied can be of
infonnation, tiireat, vulncrabilky, or affect. Alternatively, doing ooAing may
imply endurance, that is, tolerstfing a situation because it is unavoidable, one is
afraid, or one does not know what else to do. It may also imply sonte other
cognitive strategy oot di^ectaMe by chedclists of action-oriented, problem-
solving strat^es.
Little is known conceming the prevalence of the otiier intraiddly focused
responses that have been identified (detachment, illusory control, reattribution).
In an early p^qter, Gniber and Bjom (1982) found that 10% of the blue-collar
victims they studied used reattributicm as a ceding strategy, reinterfneting the
situation in such a way that it was not defined as harassment. Otbers invoked
extenuating circumstances (e.g., the harasser was lonely), or ^tempted to inter-
pr^
his int^tions as benign (Gutdc, 1985; Rabinowitz, \^0\ Se^-blame.
which we label illusory contnd, is also rarely examined, althougb Rabinowitz
(1990) m^s that it appears to be conmion. Jensen and Gutek (1982) found that
25%
of female victims attributed harassment in some way to their own behavi<a°,
an attribution tiiat inhibited botii reporting and seeking social support. In con-
trast, analogue participants are unlikely to report tiiat tiiey would blame them-
selves (Fitzgerald, 1990), anc^o- instance in which the two d^a soiaces directly
conflict.
Externally focused responses. Tbe most common prc^letn-solviag strategy
ai^iears to be avoidance; #ie litra:abire suggest that ^^{xoximately balf of em-
ployed victims activdy atten^ to cope in this way (Culbertsoov Roseofeld,
Botrth-Kewley, & Magnusson,
1993;
Ktzgerald et al., 1^8;<iiitek, 1#^; McKin-
ney, CMson. & Sattrafield, 1988; Sdmader, 1991; USMSI%, 1981; 1988). Also
common is appeasement, an aOn^ to "|Hit off" the faarassra' vnii^ut direct
confrontation (e.g., humcs", esurusra, ddayiag, ^c.).
I&IBCH:
is particidarly com-
mon ia less serious situatkms (USMSPB, 1981,19^);C&9b»-andl^pm(1982)
label tbis response "masking." In addition, 10% of tbdr blue-collar itnctims used
^laying tactics hoping the harasser wcmld "take the biat" tiiat tbjjl^ were not
mieteslted. "
Not sur{HisiBgly, a substantial numbo- of victims seek social s^ort; 68%
d&e Mexk Sy^eans (1981) vic^offi cUscussed the px^tem
mHHi
a c^oikra- and
60%
talked
sSsaat
O^r bacassaiatt wi&
faemSs
md fan^. What ^0ts teseaxdi
RespoBKs
io Baraasai«t 121
on tlie reactions of otiieis, bowever, suggests them to be less tban uni-
formly supportive (e,g.^ Loy & Stewart, 1982); thus, it is important to distin-
guish se^dng support as a strategy from the effects of support as a resource.
Women also employ a variety of assertive responses to communicate tbat harass-
ment is unwelcome, most commonly,
a
direct
request
tbat the perpetrator stop his
bebavicH- and leave the woman alone; 44% of tbe female Merit Systems <1988)
victims asked tiie bantsser to stop. In addition, 14%
tiireateoed
to expose the
barassment to others at woik, and Gmber and Bjom (1982)
reported
that
15%
of
their auto worker victims veibally attacked tbeir harassers and 7% physically
attacked tbem.
By far tbe most infrequent response is to seek instituti(mal I organizational
relief (i.e., notifying a supervisor, bringing a formal complaint, and filing a
lawsuit). Victims apparentiy tum to such strategies as a last
resort
wben all other
efforts have faOed. Not surprisingly, tbe least confrontational responses are the
most
common;
victims
are
more likely to talk witb
a
supervisor
than
file
a
formal
complaint, and legal claims are by far the least common response—^^truly, tiie
"court of tiie last resort." TTie percentage of victims who seek such relief is
influenced by a variety of factors, and it is unlikely tbat any single figure can
ever stmmiarize sucb data adequately;
frequency
of
reporting
can be expected to
rise ia tbe wake of increasing sensitivity to this issue and there is some evidence
tiiat it bas begun to do so (Wagner, 1992). Despite sucb overall increases, tiie
individual and oi^anizational determinants of such responses are unlikely to
change, and it is to tbese influences tbat we now tum.
Situational and Individual Correlates of
Coping
Strategies
Situational antecedents. Tbe most well-establisbed finding here is between
what is genraally l^xled barassment severity and assertive, extemaUy focused
responding. A variety of studies bave
reported
that explicit, repeated, and obvi-
ously barassing situations are more likely to elicit some form of assertion or a
more formal complaint (B^er & Ibrpstra, 1988; Livingston, 1982), wbereas
bebavior tbrt is transitory or less expUcit is typically ignored, met
witb
humor,
or
a\«>ided (Gruber & Bjom, 1986; Loy & Stewart, 1984). Examining the original
Meait Systems tteta (USMSPB, 1981; Livingston 1982) reported a moderate
amel^on betvi««i btoassmratt severity and assertive response, a finding repli-
cated by rtbers (Brooks & Pterot, 1991; HessMi-McInnis & Fitzgerald, 1992).
Less wcH dooimoated is tbe
relationsbip
between harasser
sttOus
and re-
sponse. Seaxt studies sug^t lbat victims
are
more l&ely to re^rt supw^sor (as
t© do-wotker) harassment (Livingston, 1982; Loy & Stewart, 1984;
, 1981). However, tiiis findUng is not unif«an (e.g., Gmber & E]om,
1982),
usd Mane have bypoftesiswd tiiat, given tiie povwr djrounics involved,
duwid respond more passively to supravisor harassment (Tbacker &
122
Ferris, 1991). A confounding factor here is
that
baiasanent by sup«>asors t^ds
to be m<Ke sevrae tiian
CO-WCHI^-
barassment (Gruber & Bjom, 1^; Loy &
Stewart, 1984); as more explicit harassment teads to dicit more assra^ve reac-
tions, any
relatioiKhip
between harasser st^us and victim rehouse is
obsaa^rf.
Finally, it seems
reasonaWe
to ejqKct
tiiat
organizational
context-^tat is tbe
organizational
norms
and culture
tiiat
are power&il
iwedictors of sexualiarassmrait
itself
(Htzgearald,
Hubn, & Drasgow,
1995;
Hesson-Mclmiis & Fitzgea-^, 1992;
Pryor, LaVite, & StoHer, 1993)—sJiould exert a cortesponcMng effecS on victim
response. Gmber and Bjom (1986) reported tiiat organizitfional factors were tbe
best predictors of response when severity of barassment was controlted (Hesson-
Mcinnis & Fitzgraald, 1992; Lach & Gwartney-Gibbs, 1993).
Indivitbtal emtecedents. In t»ntrast to situational fiactors, moa individual
variables show only yiesk relationships to victim response, one exogition being
whetiiCT a vitaim labels her experience as sexual harassment. Some evidence
indicates tiiat women wbo "feel barassed"—and tberefore label tiiek experience
as hmassment—may be more likely to respond assertively. For exan^Jle,
74%
of
womrai who reported tiiat tiiey frft barassed by ferar supervisors told tiiem to
stop,
and 12% filed a conqilaint (Saunders, 1992); bowever, botii severity and
supravistMy status are CMifounded bere witii perceptions aiKl U4»eliBg of barass-
ment. Otiier
fectews
tiiat
have been
relsrted
to
type of response inc*ide feminist
beliefs (Brooks & Penrt, 1991), sex role traditionaiity (Jensen &
QU^L,
1982),
and self-blame (Jensen & Gutek, 1982). Finally, altiiough it has been hypotiie-
sized tiiat greater vuhierability will be related to passive responding, tiiis has not
been sbown to be tiie case.
Response outcomes. The quesdon most commonly asked concerning victim
response
is,
"Why didn't she just report himT' Faced witii tiiis question, women
give a variety of answers. Tbey believe
tiiat
notiiing can or will be dboe (Gutek &
Koss,
1993; Martindale, 1^«); USMSRB, 1981), aad many
are
leluctimt to cause
pn*lems for tiie harasser (Giflek, 1985; Jensen & Gutek, 198:^ Maitindale,
1990).
Tlie
VSS3BX.
commcm reason, bowev», is fear—fear of veisSsakm, of not
being beMeved, of hmiing one's career, or of beii^ shamed aUd fcunuliated
(Frtzgei^d et al., 1988; Gniber & Bjom, 1982; Gotdc, 1985; Gattk & Koss,
1993;
MM^Mfede, 1990; Hafflips et al., 1989; Saondos, 1992; USIlSSPB, 198i,
15«8).
Unforttmately, sucb beHds are often well founded.
Sixty-two perceia of tbe stjtte wrakers ^iKiied by Loy and SBewart (1984)
repeated
n^iaticm
for tiieir responses to harassmrait, inchidi^^awered job
evaluatioiis, denial of
jMXMmrtion,
aod beiag transferaed at firad—|iBd the most
ass«»fiveTO^onses often incurred the ^e^est coids. AMiou^ Ad^MBple size
was
OTiafl
(/V = «6), tins ^tftem is coBsiamt ^((^oflier tesearcbd^iti
Sesptmses to flarassmett 123
link between more assertive responding and negative outcomes (Culbertson et
al.,
1992; Hesson-Mclnnis & Fitzgerald, 1992; Livingston, 1982; USMSPB,
198 J, 1988). For example, one-tbird of tbe Merit Systems (1981) victims wbo
filed fonnal claims said tiiat it "made things worse." In Culbertson et al.'s (1992)
s&idy of tbe Navy, one-tbird of the victims said tiiat, when they complained, they
were humiliated in front of otiiers and Hesson-Mclnnis and Fitzgerald (1992)
found tbat assertive responding was associated with more negative outcomes of
every type (job, psychological and health-related), even after severity' of harass-
ment was controlled.
Several studies examining outcomes of legal action—the rarest of all
responses—find tiiat plaintiffs typically do not fare well in court. It is estimated
that only about
1 %
of victims ever take legal action, and the percentage who win
such cases has been found to range from one-third (Terpstra & Baker, 1988) to
one-half (USMSPB, 1981). Wben plaintiffs do prevail, settlements bave typ-
ically been quite small. Terpstra and Baker (1988) reported that in cases filed
witb &e Illinois Department of Human Rights, the average settlement was
around $3(XX). Finally, such assertive action often incurs considerable cost. In a
review of cases filed in California, Coles (1986) found that half the victims lost
their jobs and an additional 25% quit in fear and frustration.
In sum, despite pervasive public opinion tbat women should "handle" ha-
rassment assertively, confront tbe perpetrator immediately, and repKJrt him to
appatapn&te
authorities, reactions to such responses are generally not favorable
for tiiose who actually "blow the whistle." Furthermore, taking the matter to
court is a ridcy and often traumatic venture (Dolkart, 1992). As Livingston
(1982) states, "Given the immense psychological and economic costs to individ-
uals who use formal action, in contrast to the potentially meager gains, it is not
surprising that so few victims choose tbis response" (p. 15).
Critique and Theoretical Framework
Our review of tiie victim response literature reveals it to be atheoretical and
dominated by sioiplistic notions about the effectiveness of assertive responses.
Ws prc^xjse to reconcxptualize responses within the context of modem cognitive
apfffoaches to understanding stressful life situations (Carver, Scheir, & Weintraub,
1989;
Lazams & Fbficman, 1984). Sucb an approach allows us to constmct a
psydiological foniMilation of sexual harassment (as opposed to a purely legal
(me) and to invoke the extensive body of stress and coping research to analjrze
re^xmses to barassment. In addition, it provides a psycbological explanation of
gendei' differences in perceptions of barassment, resolves tbe conflict between
analq^K studies and tiiose of victim behavior, and provides a more
frame fcff tiie queiy, "Wbat is an effective response?"
124
Sexual Harassment as a S^vs^itl L^e Event
From the cognitive perspective, '^ydicdogical stress . . . is a
between the person and
the
^lvironmrait
tiiat
is
{^raised
by
tiie
prason
as
or exceedii^ bis «- her resources and endangering his or her well-beii^" (La-
zams & Fblknum, 1984, p. 21). Stimuli typically experieoe^^ stisssiul include
botb major life evems (e.g., divorce), and tbe more fr«pesDt but tess drastic
hassles (e.g., time pressures, inteipersonal difficultjes (Kamier, Coyne, Scb^-
fer, &. Lazarus,
1981)].
This distinction maps smootiily onto the<domam of
sexual barassmeait, in wbicb many experiences are of tiie latter type ^.g., offen-
sive comments, annoying ^tention of
a
socual nature), l»it
an
important minority
resemble tbe fomier (e.g., sexual coercion, assaultive experiences, aad tbe like).
Importaatiy, severity of tie stressor is not considered to inhere in tbe event
itself;
ratiio; it is si individual's evaluation of tbe situation, as influenced by factors
sucb as ambiguity, perceived threat, loss, and so fcxtii, tbat is detenninative.
Conceptualizing harassment in tbis
mann^
bas sevcafsi advantages. First, it
pirovides a psychological perspective m barassm«it consistent
witii
l)ut broader
than the legal (tefinition, tbus allowing Ux si&iatioos tbat are cleaily barassii^
within the generally understood m^uiing of tiie term but unlikely to meet legal
criteria for sta&itoiy vio]ati(Hi. Hie range of psydiologically noxious woil^lace
situations wkh whidi women are amfranted is cleariy broads: thai those pro-
soribed by legal formulations, and any meaningful psydiologicd tbeoiy of sexu-
al harassment must be aUe to (fisentangle the psycbok>gicaJ consteact from tbe
legal one. At the same time, if tiiis research is to ctmtribute to sexuji barassment
jurispmdence, it must be {^rationalized in ways consistent witb thedegal defini-
tion,
but that allow for the distinction betwera wlutf is offensive
^and
wbat is
illegal. Only the latter is a matter
for
judges and juries, but botb are a{qnopriate
topics for psycbological soutiny. A stress and coping perspective fits tiiese
criteria nicely.
Second, tbis framewoik avoids a naively stimulus-based foroudation tbat
implies certain types of brassing situations me inbraentiy or Bonoatively more
sbessful (i.e., "severe") tbanotiiras, aad acknowledges tiie role of^eQgnitions ia
detaxoining individual phenomraological e^qierience. In contrast^t (^ existing
liteiatuie on sexu^ harassm^fflt coi^sientiy invokes a stimulus-bsised severity
c«itiiBium, beginnmg with Till's (1980) five "levels" <rf'bacassmott^md Merit
Systems, (1981,1988) (Mstinctionbetweoi less severe and nuHesevarebs^uadons.
not denyii^ Aot cettmi experiences are oifarjectively m^ e^egious
otiicfs, we
a^gue
^at sevmty does not inbrae scjely in
tiiej^aiiiBiUus
act
kseif,
at least m tbe amjori^ ctf cases. Such tiiii^ as fin(|uency,
i^tcaSaoa,
and
peicei%%d
oSssmvesx)s&,
as well as indivitbial iustoR (e.g., pcevi(||t vic^mza-
tioa. peiceived
iOsteal,
economic vuinefabiUty) wt involved ia
aa^
^iiSsBawaslaaa
of tbe psycbologica] severity of a barassiog siniaticm. Aad, wbate^ gradations
Besponses to Haramnent 125
of stimulus intensity do exist are not well captured by simple classification of
events as less or mote serious. After all, the behavicM^ at issue in Robinson v.
Jacksonville Shipyards (1990; e.g., a long-term barrage of offensive posters,
insulting comments, pu^owns) would be classified at the lowest level in the Till
system and "less serious" by Merit Systems, although the extended pattern of
behaviors involved are severe by any criteria—an evaluation witii which the
Robinson court agreed.
The Role of Cognitive Appraisal
"Cognitive appraisal {is] the process of categorizing an encounter, and its
various facets with respect to its significance for well being. It is not infonnation
processing, per se ... [B]ut rather largely evaluative, focused on meaning and
significance, and takes place continuously during waking life" (Lazarus & Folk-
man, 1984, p. 31). Tbe process is typically dichotomized into primary and
secondary appraisals, the former concemed with whether a stimulus is personally
irrelevant, benign, or stressful, and the latter posing the question of response—
that is, wbat to do?
Primary appraisal. Research on individual differences in perceptions of
barassment bighlights the utility of the primary appraisal concept for understand-
ing responses to sexually harassing situations. Far too extensive to examine here,
this literature suggests the existence of gender differences in evaluation of po-
tentially harassing behaviors, a gap that is particularly lai^e with respect to
hostile environment situations. Women have been found to be more likely than
men to view such situations as offensive and threatening in virtually every study
ever conducted (see Frazier, Cochran, & Olson, this issue). Such differences—
intuitively reasonable but so far untheorized—are readily interpretable from a
stress and coping perspective, which posits that a stimulus will be evaluated as
stressful to the degree that it represents harm, loss, or the threat of harm or loss.
That women as a group are likely to appraise uninvited sexual behavior as
threatening, whereas men view it as either benign or irrelevant, was intuitively
understood by tbe judge in Ellison, when be proposed for tbe first time what has
come to be known as the reasonable woman standard: "Because women are
disproportioiuuely victims of rape and sexual assault, women have a stronger
incentive to be ooncemed witb sexual bebavior. . . . Men, wbo are rarely victims
of sexual assauk, may view sexual conduct in a vacuum without a full apprecia-
tion of tbe social setting or tbe underlying tbreat of violence that a woman may
perceive" (ElUson v. Brady, 1991, p. 879).
Similarly, tbe conflict between analogue and >dctini studies is resolvable
tbe concept of |Mimary i^praisal, as tbe usual procedures of analogue
(e.g., p^)er-and-pendl scenarios, videos) will not invoke a stiess re-
126 ntzgeraM et al.
sfxmse as participants are unlikely to view tbe stimulus as 3<^)ncsenting mty tbreat
of hann or kiss to them personally, hideed, tbis is one of tbe ^incipal reasons for
employing sucb ]HX)cedures to study {Aenomena otiienvise not amenable to ^bi-
cal experimentation. Analogue (nocedures are by definition nonsttessful said tiius
bave little to say about tbe experiences and reactions of actual vicdms, are likely
to be substantiveiy misleading, and are vulnerable to being exploited for inap-
propriate purposes. As we see below, responses to any situfttion we intimately
connected witb tbe evaluation of that situation as benign or tbieatening; to tbe
degree we ignore this connection, we risk producing both bad science and bad law.
Secondary appraisal. Secondary appraisal is the psycholo^cal process of
deciding bow to respond to a siau^on perceived as stressful, a decision tbat
dqiends on what options are realistically available as well as whitt is at stake. "It
is a complex evaluative process th^ takes into account wbicb co^g options are
available, tiie likelibood tiiat a given option will accomplisb wbat it is supposed
to,
and the likelihood tbat one can apply a particular strategy or set of strategies
effectively" (Lazams & Foikman, 1984, p. 35). Tbis concept provides powerful
insigbt iato victim respcxises to sexual harassment. For example, research con-
firms that victims frequently do not report harassment because tiiey believe it is
useless <x dangerous to do so (Culbertson et al., 1992; Fitzgerdd et al., 1988;
Gutek, 1985); Gutek and Koss (1993) implicitiy invcAe this coacqpt when they
suggest tiiat "women may perceive the direct methods [of respcmding] as riskier
and less certain in tiieir outcomes tban tbe indirect metiiods" ^. 40, empbasis
added). In a direct test of tbe importance of outcome expectations, Ormerod
(1991) found tiiat fear of negative outcomes was the most powerful predictor of
responses to harassment. Only in situations involving explicit sexual coercion did
ti^ expectation of stopping tiie behavior outweigh the fear of ret^iation. As we
bave seen, such fears are frequentiy realistic. In genea^, the research supports
the importance of c(^iitive a{^[H:aisal for influencing victim res|>«ise. With tiiis
background, we pn^se in tbe next section
a
cognitive theoretidU
friamework
for
understanding reqioDses to sexual barasanent. •
Wilb Sexaai Harassment: A Theoretic^ Fraaework
As we bave seen, tbe literafiire on reqxmses to sexual barassment bas
focused afanost exclusively on {nt^lran-focused strategies, discussing tbem pri-
marily in terms of nmsx^ (i.e., bow effective tbey were in stop|plBg barassmoot)
aed in^lkitiy involdi^ a trait perspective by asking wbat tbe pHnid{nnt woi^
do or tiie victim ^umld do in a vari^ oS ciicumstances. In copirast, tiie stiess
aid
a^Big
litra^diiie wsexts "acting (icpresra^) coimantiy dinging cognitive
Md betevkxal eifocts to man^e spedfic &aacaid and/or
naetif^
demands ibat
ate ^ifxaised ws taxing or racceeding tiie resoucces of tbe pa^^" (Lazanis &
Responses to Harassment 127
Folkman, 1984, p. 178). Ceding is thus defined as a process ratber than a trait,
concemed witii what is actually said or done in a specific situation. According to
tbis perspective, coping has a dual function: (1) to manage or alter tbe distressing
situation itself {problem-focused coping) and (2) to regulate emotional reaction
{emotion-foci4sed coping). Explicitly rejected is tbe equating of coping with
mastery; rather, coping is defined as all efforts to manage a stressor, unsuccessful
as well as successful, thus avoiding confounding the process of coping witb its
outcome.
Coping as an Interactive Person-Environment Process
This cognitive-phenomenological approach identifies a number of factors
that influence behavior in any particular situation, including the appraisal pro-
cess,
personal resources (e.g., beliefs, commitments, behavioral skills), and
personal and environmental constraints (Lazams & Folkman, 1984). Tbis latter
notion is particularly helpful for understanding the outwardly passive responses
frequentiy displayed by victims of sexual harassment. Personal constraints on
coping include not only individual psychological characteristics, but also "inter-
nalized cultural values and beliefs that proscribe certain ways of behaving"
(Lazams & Folkman, 1984, p. 165), a clearly important influence on women's
behavior in a gender-schematic society. In one study, women with hi^ guilt over
assertiveness reported being less assertive in social contexts (Klass, 1981), and
Gutek (1985) reported that an important inhibitor of reporting harassment was
not wanting to hurt tbe man involved—clearly a gender-linked cultural value.
Even more powerful are environmental constraints on victim behavior, par-
ticularly organizational practices that thwart coping efforts. We have previously
reviewed evidence suggesting that assertive, problem-focused responses to ha-
rassment are affected by organizational constraints, what Hulin (1993) has la-
beled the organizational climate for harassment. Additional data are provided by
Gutek (1985), who reported that 60% of the nonreporters in her sample believed
they would be blamed for the incident if they made a formal complaint; another
60%
believed complabits would be ineffective because notiiing would be done.
Studies of victims consistently report that fear of personal or organizational
retaliation is tiie major constraint on assertive responding. The failure to incorpo-
rate sucb powerful extemal constrwnts is a major flaw in tiie analogue researcb in
tiiis area wA in tiie victim response literature more generaUy.
Finally, tiie notion of coping as a process, ratber tban a trait or a singular
event, allows far cbanges in cognitions and bdiavior as tiie situation unfolds. As
Gutek and Ko% (1993) point out, harassment is generally not a one-time event
but naher unfolds across time, as do victims' responses. "How a woman re-
spraids probifely depends on tbe progression of tiie barassment. Has sbe riready
tried ignOTii^, eVadmg, or joking? Is tiie harassment continuing, escdating.
128
becoming more hostile or
tbreatening?"
(Gutek & Koss, 1993, p. 41). Sriisbury,
Ginorio, Remidc, and Stringer (1986) describe progressive em<^onal reactions
to harassment among victims in tiieirapy (coniiisi(»i/sell-blame; fear/anxiety;
depression«'anger; disillusionment) &at are likely linked to rea|}praisal of tbe
evolving situation and mediated by organizational responses.
Emotion-Focused
and
Problem-Focused
Coping
Atten^ts to examine coping witb sexually barassing situations bave mainly
identified problem-focused strategies, that is, attempts to manage or alter tbe
situation in tiie extemal world. Partly, this is an artifact of tbe manner in wbicb
tbe research is framed or operationalized. Studies tbat bave surveyed actual
victims (Culbertson etai.,
1992;
Gutek, 1985;Martindale,
1990;
USMSPB, 1981,
1988) typically present them with a brief checklist of action-oriented strategies,
tbus eliminating the possibility of identifying more emotion-focused responses.
Analogue studies
are mcwe
likely to utilize an open-ended strategy (e.g., Terpstra
& Baker, 1989), but still
"pull"
for a problem-focused response by asking, "What
would you/should she tio in this simation?"
Tbus,
we bave a fairly good picture of tiie range of problem-focused strate-
gies,
as well as considerable data telling us tbat
real
victims do aot use tiiem very
often. OthCTcategories available in most studies (e.g., ignoredJhe behavior; did
nothing; went along with it) are in general simple negations ®f action-oriented
stiat^ies and tell us littie about bow endorsers actually attempted to manage the
situation psychologically. As we noted, doing nothing may
repeesent
a variety of
things (e.g., a cognitive
aa^im&al
suggesting tiiat the situation is nonthreatening
at irrelevant; a deliberate attempt at extinction; endurance ba^ on fear or the
belief tiiat tbere is no effective option available); equally ii^xHtant, it likely
masks a variety of
attenqjts
to manage emotions and cognitions :^ut tiie experi-
eaoe. Tbere is a deartii of infomiation on sucb strategies despite mformal indica-
tions tbat tiiey are ubiquitous.
A Cognitive Conceptualization of
Victim
Responses to Sexual Marassment
Given tbis analysis, we can now outiine a framewoik for und^standing
victim responses ooosisteat witii a xaxe general cc^iitive ftiess and coping
perspective. Sexual barassment can be defined psychologicalt^as unwanted sex-
rekted bebavior
at
woik tb^ is appraised by
tiie
recipient
as ofietasive, exceeding
hea-
resources,
or
tiusatening
b^ well-being. Coping is tiie prooi»s of isspoading
to sudii unwanted sex-iekued bdiavlor and includes botb practical aoen^ to
solve, maaage, m
altea-
^ae
situation (^OeniaUy focused co{ni!||and atteo^ to
co§nMve and emotional reactions to
tb&
utaaAon ^BSemeSy focused
Bespoascs to Haraasmoit 129
Hie way in wlucb any individual will cope with potenti^y barassing si&a-
tioas depends on (1) ber cognitive evaluation of tbe situaticm with respect to its
significance for weli-being (i.e., is it irrelevant, benign, or tbreatening) and (2)
tiie <^ons tbat are
realistically
available, tbeir costs and benefits, and wbat is at
stake. Sudi evaluations are part of a complex, reflexive {nocess tiiat cb^mges
over time as tbe situation unfolds. Additional influences on response include
lesouroes and constraints, botii personal and environmental; tbese include botii
individual characteristics (e.g., assertiveness, sex role attibides, economic and
psychological vulnerability) and also organizational ones (e.g., climate, toler-
ance for barassment, etc.)
Rom tiiis perspective, it is not possible to assert a priori that any particular
category of response is, by defimtion, eitber appropriate
or
effective; we specifi-
cally reject tbe equation of assertive responding with effectiveness, as well as
that of coping with mastery more generally. Coping is a psychological process
that can lead to a variety of outcomes, both psychological and practical, good
and bad, depending on a large number of factors. The issue of what
responses
are
associated with what outcomes under what conditions is an empirical question,
ratiier tban an a priori assertion. Continued emphasis on tbe antecedents and
facilitatcns of organizational complaints and legal action, to tiie neglect of otber
reactions, can only
result
in a nanow and atheoretical body of
research
and lends
scientific ciwMbility to legal strategies and decisions that blame the woman fw
her own victimization. It is to this topic that we now tum.
Legal and Judicial Constnictioiis of Women's Responses
to Sexual Harassment
tt)he gFEtvameo of any sexual haiassment claim is that flje alleged advances were 'un-
welcome'. . . . The correct inquiry is whether respondent by her conduct indicated that
the alleged sexual advances were 'unwelcome'. . . "
Meritor Savmgs Bank v. Vinson, p. 2406, emphasis added
Although
research
reveals
the impossibility of predicting how any particular
woman will react to sexual harassment, juries and appellate courts have consis-
testiy construed women's behavior to mean that tiie barassment was welcome,
did not occur, or could n<H bave been tbat bad—interpretations referred to,
respectivdy, as tia Slut, Nut, (or So Wbsrt?) defenses (Estricb, 1991). As witii
rape, wtdcb Estricb notes is tbe only crime "defined more by tiie actions, reac-
tions,
motives and inadequacies of tbe victim tiian by tbose of tbe defendant"
(1991,
p. 815), sexual barassment cases are frequently characterized by a pre-
dominffltt focas
<m
vktim—as o{qx>sed to perpetrator—behaviw. "(R)ules and
ptx^u<Sces haw beai brarowed almost wboleside from traditional rape law. Tbe
focusim tte conduct of tiie woman—^ber reactions or lack of tban,
brar
resistance
OT lade of k-'iai^esR witii only tiie most
minor
cban^s"
(p. *15). Tbas, it is
t30
instructive to contrast sexual bwassment jurisprudence witii tiie legd btstory of
o&er fwms
<A
Titie vn barassment (e^., cace). Akbou^ rec^it decisions
conceming race, natiOTial origin, and
religious
baiassramit eacb li^ "unwelcome-
ness"
as an impcHtant element (e.g., Boutros
v.
Omum Regioiud Dxmsit Authori-
ty. 1993; Meltebeke v. Labor and Industries, 1993;
Vrnice
v. ShuOiem
Bell,
1989),
not one of tbese dedsions tneatioiMd, ev«i in passing,
wheSio-
<»-bow
&e
fdmntiff d^nonstrated tiiat the barassment was unwelcome; ra&er, tbey {Me-
sumed tbat tbe egregious iratuie of tbe barasser's conduct was proof enougb tbat
it was intrinsically unwelcome.
Hie decisicm io
Vance
is paiticularly telling. Tbe only Blacic employed in
bis defiartment, Vance wie repezrtedly bioassed: An effigy was suspended iix>m
his doorway at woik aad bathroom graffiti read "KKK." Tbe court reasoned.
There are difieieiit ways in which a {daintiff initially responds to or copes with harass-
ment. The employee may quit 'cold turkey' or, as in {Ptsmtiff's] case, may expetiraice a
prolonged period of turmoil is which die individual regularly avoids Ihe workplace for
feu-
ol su&ring furtho- hanssment, Aheraatively, an employee may reaa«ngrily to the
racial hostility, and may nxnc easily be ptovofced into arguments or (AysitaA altetca^ns
widi tfiose co-wotken responsible, , , , (p, 1506)
The court also took particular care to ocplain Ae vi(^m's silence in tbe face of
bis ordeal, wbich extended over a period of ten years. Tbe ki^age of tbe
(^iinion valorizes bis cbaractra': "[be] was tbe victim of racial tanitts sucb as tbe
nickname 'Buckwt^at' for the first ten years diat be was employed. ... He
asked bis cowoik^s to back o£f, but, for tbe most pait, be cbose to tum
Ihe
otber
cheek and ^odure tbe prejudice." In essence, tbe court focused not
<HI
wbetber
barassment was welcome, but lad^ wbetber Vance's reactions substantiated bis
claim that tbe events bad occurred, tbat is, bis credibility. Analyses of sexual
barassment victims aie dramaticaily different; not ooly do couitsemploy a differ-
ent focus (tb^ ctf welcomeness), bttt also view tbe bdiavior tbroug^ a decidedly
less syn^tatbetic lens.
Welcomeness
and the Legal Construction of
Victim
Roles
As corate stnig^ to analyze vktim bebavior as an index of wbetber die
aSegied
conduct was welcome, diey freqii^ndy (»igeonbote viotiais into one of
sevezal ndes, focusmg oo selected aspects of tteir actions ani ^miltaneously
ignoting (^lers. For exai^iie, tte Silem Tblerator's failure to ngssk aal is con-
strued as wekxxnkig ifae xtvances or conmaents, «4ieieas tbe piaeticqiation of tbe
Instiga^r-m-Kind in various activities is seen as oonsem to iMsaai beba^^or.
Sacfa l^ally canstiacted roles neveal a gent 4ieai about bow
pi^)fex
barassment
victims ate expected to b^ave, as
leal victims actually do bebave.
meoL,
mhg^js^oDsee to battiraed women (e.g., see FisdMT,.|gda]a£,
131
1993) and rape survivors (e.g., see Fischer, 1989) reflect similar misunderstand-
ing of women's experiences of male violence.
The silent tolerator. One of the most common sets of facts in sexual barass-
ment litigation involves a situation where a woman endures or tolerates harass-
ment for some period of time, never speaking out publicly or reporting to man-
agement before leaving ber job. Unlike the racial harassment decision cited
above, courts bave been consistently inbospitable to sucb bebavior, typically
interpreting wbat we bave labeled "endurance" as consent and burdening die
victim with proving that ber conduct was proper, ratber than tbat the barasser's
was improper (Estricb, 1991). A number of coiuts bave found that inaction or
silence on the pan of the victim is, as a matter of law, an indication that she
welcomed die harasser's conduct, A pertinent example involves the case of a
student intern who worked in Minnesota county govemment, during which time
ber supervisor made rejteated advances to ber (Kresko v. Rutti, 1988). Tbe
Minnesota Appellate Court found that the fact that she did not complain was
dispositive tbat ber supervisor's bebavior was welcome, A more dramatic exam-
ple involves a Tennessee juiy that declined to convict a judge who took advan-
tage of the privacy afforded by the bench to fondle the legs and genital area of the
female clerk wbo sat beside bim during trial. Choosing to discount toe fact tbat
tbe cleric had complained immediately and vigorously to her supervisor, tbe jury
reasoned tbat by not reacting overtly in the crowded court room, the clerk had not
given sufficient indication that the conduct was unwelcome (U.S. v. Lanier,
1992).
Tbus, botb judges and juries clearly expect that a real victim will speak
up,
botb publicly and privately, creating tbe image of an idealized woman wbo
bebaves in the proper, socially constructed way in the face of offensive conduct.
As Estricb (1991) summarizes.
Such a woman is tough, not 'hypersensitive"; she is aggressive, not passive. Such a
woman complains in a way that effectively stops the harassment. Such a woman does not
si^er in sil^ice or c<mfide only in other women. In short, the reasonable woman is very
much a man, (p, 846)
The instigator-in-ldnd. Intol^ant as the law may be of silence, it bas even
less sym{»>tby for tbose who cope by "going along with" tbe situation (i.e., wbat
we bave identified as a form of appeasement). The most illustrative case in tbis
area is Reed
v.
Shephard (1991). Reed,
a
jailer for tbe county, was handcuffed to
tbe toilet, tbe dnink tank, and inside tbe elevator; made the subject of lewd jtdces
and Kanaiks; bad ber bead forcefully sboved in ber co-woikers' laps and an
electtic caltk prod forced between ber legs—facts that were not disputed by tbe
defense. Hie couit
fey any objective s&mdanl, the behavior of the male deputies and jailere towaid {flie
plaintiff:] , , . was, lo say Ihe least, rqjulsive. But apparently not to Iher] . . . (JShel aot
132
oaiy eiqierienoed lius
dqpravi^ wilh amaang resilience,
ing
in kind . . . fDhe oonclusion , . , must be that [she] participated fieedy in nuny of
these
antics aad in iiact instigated some of them. (p. 489)
Tbe court knew Reed enjoyed tbese incidents, tb^ daimed, because ^
was koowa to bave used offemive language on tbe job and wore^a
T-^ikt
to work
witbout a t»a (an "exbibiticmistic babit," daiised tbe court). In addition, dte
participated in "suggestive" gift giving to male co-woikns
and
"enjoyed exbi^t-
ing" a surgical scar on ber abdomen. Tbe court aoied tbat^eed must bave
wdcomed b«' e^qieriences because otiier female officers were mt (known to be)
barassed. Chie wommi testified tbat sbe bad successfully detected sexual joking
by askii^ tbe offenders to stop; Reed, on tbe otber
iamd,
bad sot comfrfained or
asserted borself to b^ co-w<Mkeis. As bo' testimony reveals, ^ viewed berself
as a Silent Toterat(»'. Ho- p^ncqjtion was tbat sbe had to paiidcipate to be ac-
cepted by ber co-woikns: "It was impoitant for me to be a p^ce officer and if
tbat was tiae only way Aat I could be Kcepted, I would just put up widi it and
kqit my m«itb
sbut"
(p. 492). Tbere was also tbe issue of pbysical saf^: "One
dung you don't do as a police officer, you don't aiitcb out am^er police officer.
You could get burt" (p. 492).
WidKwt tbe full record of die trial, it is impmsible to determine wbetber
wbat was constructed as "instigation" (i.e., sbe i^ked for it) was in reality
anvival; bowever, it is not difficult to understand diat a woman mig^ see sudi
bdiavin- as oecesssy in die macbo environing of law ^ifarceinent. Even
absoit tbis framewoik, we find it difficult to undrastand tbe cooit's conclusion
tbiu going brale8s to work meant de jdaintiff enjoyed baving a cattle {xod sboved
between bex legs. It ^xxs not i^pear tbat ba bebavior was tndy "in kind"—sbe
did not handeuff
hn-
co-woricerB to tbe toil^, or sbove tbek beads into ber
lap.
As
SUianey (1986)
remarked
in an analysis of
an
earlira'
case "flli&couits concluded]
tbe plamdffe consented because tbey engaged in acts similar 4o
tibe
ones about
wbicb they cmx^ained. But die plaintiffs' actions were not tbe acts about wbicb
tbey Mwnplainfid; to suggest tbey are is to focus attention oo tbe plaintiffs'
conduct, not tbe baiassers' conduct, leaving unexanuned die liaassers' respon-
sibility for dieir acts" (p. 1134).
Coiscnt
m-
Cvfli^? Toward an I^i^aaded U^n'slaiidiBg
«r
Wdcmneness
in SoEoai mnuaaaeatt Cases
As stnictared by die Vinson decisitm, die comai iaq^iy of any sexual
barassment claim is welcomeness. Uidike cases (tf radal an<|^Kligious barass-
waA,
Ddieie aSensiw beiamorisassumedto be unweicrane, liis Stqmoie Court
bas placed wetetmteness at tbe beait ci ^x sexual btawamoi^ttqfmf aad buF-
deaed die victka w^ pBovk^ die
a^iier
invited
SMH'wdoonpl
tbe cosduct
<rf
wbidi die is complaiimig. iWftet, comts tenv con^ndly impied on dieir own
RespoiiKs to Harassment 133
particulmzed constructions of victim behavior and repeatedly equated coping
witb consent.
Especially problematic is the pattem of putting victims in a double, or even
triple, bind. Silence equals consent—but so does speaking out. In the ultimate
irony, one court went so far as to equate speaking out witb silence! In Ukarish v.
Magnesium Elektron (1983), the DC, Circuit concluded that aside from com-
plaining to her siq}ervisor and keeping a diary detailing her abuse, the plaintiff
"appeared to accept (woricing conditions] and joined in [them] as one of tbe boys
and did not complain to anyone" (p, 1319, empbasis added). Playing along with
tbe siaiation implies a willing participant, pardcularly when sbe is silent as
well—and evidently even when she is not. Despite the Vinson court's direction
to consider the "totality of the circumstances," cotirts repeatedly focus dispropor-
tionately on the victim's behavior, ccmstructing it in a particular way and ignor-
ing other, equally likely, meanings. As long as the welcomeness inquiry remains
central to Tide VH claims, women's behavior will continue to be scrutinized and
found wanting, much as women's "resistance" remains the focus of rape trials
even today, Tbis has led some to suggest eliminating welcomeness as an element
of harassment claims altogether.
Suggestions tbat the communication burden be shifted from tbe plaintiff can
be found in tbe legal literature for over a decade (Estrich, 1991; Shaney, 1986;
Weiner, 1983). In an early note, Shaney (1986) proposed tbat an appropriate
standard would require that courts (1) first focus on whether the victim overtly
indicated consent and (2) then examine whether such consent was fireely given.
As she notes, sucb an analysis closely parallels the Supreme Court's analysis of
Mechelle Vinson's behavior. The Court concluded that although Vinson's sexual
relationsbip with her supervisor was voluntary, it was not consensual, i.e., they
recognized tbe importance of coercion in determining her bebavior. Sbaney
contrasts this distinction to a series of lower court decisions equating sUence
(endurance and otber intemally focused responses) and participation (appease-
ment) with consent. She argues, "Such a presumption is flawed. It fails first to
assess the defendant's conduct, fails further in not recognizing the possible
coercive nature of tbe defendant's conduct, and fails finally because it does not
examine tbe effects of coercion on the plaintiff's responses" (p. 1128). From a
psycbological perspective, her reasoning is flawless.
Like Sbaney (1986), we propose a two-pronged test, tbat is, presence of
consent phisftxedom of consent; bowever, we also propose two additional con-
siderations. First, we suggest distinguishing between the two forms of hostile
environment, tbat is gender harassment and unwataed sexual attention (Gelfcuid,
Fitzgerald, & Drasgow, 1995). As recendy emphasized by the Equal Enjoy-
ment (^poctunity Commission (EEOC, 1993X gender harassment is "verbal or
physical conduct diat denigrates or shows bosdlity or aversion" (p. 51269);
examples include epidiets, slurs, taunts, and gestures (e.g., Rabidue
v.
Osceoia
134 fltzgeraia et al.
Refining. 1984); the display or distribution of obscene or pornographic materials
{e.g., Robinson
v.
Jacksonville Shipyards, 1991); gender-based hazing (ffwn'j v.
Forklifi Systems, Inc., 1993); and otiier threatening, intimidating, and hostile
acts (Reed v. Shephard, 1991). As with racial harassment, to whicb it bears
peibaps the closest resemblance, such behavior should be presumed to be un-
welcome; the appropriate inqaii7 tbus becomes (1) whether the bebavior actually
occurred, and (2) whetber it rises to the level necessary to trigger the standard.
Victim response will of course still be relevant to such inquiry (e.g., possibly she
didn't report him because she originally considered the behavior *ivial or non-
threatening); however, tbe focus of the inquiry shifts to her pereeption of his
behavior (i.e., primary appraisal), rather than her response to hist>ehavior (i.e.,
secondary appraisal).
Widi resjject to unwanted sexual attention, the issue is somewhat more
complicated, as consensual sexual relationships do of course occur in the work-
place, Hius, we acknowledge that consent can be a legitimate area of inquiry and
proof.
However, with due respect to the Vinson court, we argue that it is not
possible to unambiguously infer consent solely by citing the responses that a
woman makes to sexual attention in the workplace. Victims react in a variety of
ways for a variety of reasons; in particular, the choice not to file a complaint
appears to have mainly to do with a well-founded fear of retaliation and cannot
be taken to imply that the behavior was either welcome or trivial. There is simply
no typical, usual behavior following sexual assault (Fischer, 1989). Further,
burdening the victim with proving that the conduct is unwelcome assumes that
sexual advances by any man to any woman are by definition welcome undl she
proves otherwise. We submit that this is a very odd assumption. Indeed, it seems
to us far more reasonable to suggest the opposite: that sexual attendon in the
woricplace should be presumed to be unwelcome, unless the initiator determines
otherwise. Thus, the burden falls on the defendant to demonstrate how he knew
he was welcome. Shifdng the burden of communicadon in this manner focuses
the inquiry where it belongs—on the defendant's behavior, rather than the plain-
tiff's.
Summary and Conclusion
Fitzgerald and Shullman (1993) note that researeh on womea'* responses to
sexual harassment is an important enierging trend in the literaturein this area. As
our review reveals, such research has been mainly atheoredcal in-nature and, by
its preoccupation with reporting and filing complaints, lends scieilffic credibUity
to tbe courts' misguided insistence diat silence—or at least, R0t reporting—
equ^ amsent. Tbis problem is magnMied by the overdependenie on analogue
methodologies aoA available samples of nonvictims; unlike c^ege students
speculating ^x)ut hypmtbetical scenarios in a safe, nonthreateiSag classroom.
Responses to Harassment 135
real victims respond in a variety of ways that the courts choose to discount. In
addition to asking or telling the harasser to stop (as Anita Hill testified that she
repeatedly did) real victims avoid both the perpsetrators' presence and their impor-
tunities; they offer rationales and excuses for why they cannot comply; they joke
and evade, and most frequently of all, they simply endure, hoping that the
situation will eventually go away without the embarrassment and retaliation that
so often accompany a formal complaint. Inhibited by socialized personal and
environmental constraints, many cope with the situation solely through psycho-
logical means, telling themselves it is not really important, or that he did not
mean it, or it must be their fault.
If psychological research is not to be misused, researchers must begin to
assist the courts to undentand that the equation of such reactions with acquies-
cence, welcomeness, or consent is simply not justifiable. Further, if eitiier wel-
comeness or resistance is to have any meaning, interventions must be developed
to ensure that organizations provide a safe environment for victims to express
that resistance through both formal and informal channels. Training programs
designed to teach women to be more assertive and to access organizational
complaint procedures, well-intentioned though they may be, are insufficient and
misguided, without meaningful organizational protection from stigma and retal-
iation. Like Perlin and Schooler (1978) before us, we argue
There are important human problems, such as those diat we have seen in occupations, that
are not responsive to individual coping responses. Coping with these may require inter-
ventions by collectivities rather than by individuals. Many of the problems stemming from
arrangements deeply rooted in social and economic organizations may exert a powerful
effect on personal life but be impervious to personal efForts to change them. . . . Coping
failures, therefore, do not necessarily reflect the shortcomings of individuals; in a real
sense, they may represent the failure of social systems in which individuals are enmeshed,
(p.
18)
References
Baker, D. D., & Terpstra, D. E. (1988). Determinants of reactions to sexual harassment. Paper
presented at the 48th Annual Meeting of the Academy of Management, Anaheim, CA.
Boutros V. Canton Regional Transit Authority, Westlaw #229718 (6th circuit Ohio 1993).
Breznitz, S. (1983). The seven kinds of denial. In S. Breznitz (Ed.), The denial of slnss (pp. 257-
280).
New York: Intemational Universities Press.
Brooks, L., & Perot, A. R. (1991). Reporting sexual harassment: Exploring a predictive model.
Psychology (^
Women
Quarterly, 15, 31-47.
Carver, C. S., Scheier, M. F., & Weintraub, J. K. (1989). Assessing coping strategies: A theoreti-
cally based approach. Journal (^Personality ami Social Psychology, 56, 267-283.
Coles,
F. S. (1986). Forced to quit: Sexual harassinent complaints and agency response. Sex Roles,
14, 81-95.
Culbertson, A. L., Rosenfeld, P., BooCh-Kewley, S., & Magnusson, P. (1992). Assessment of sexual
harassment in the Navy: ResuUs of the 1989 Navy.wide survey (TR-92-11). San Diego, CA:
Navy Personnel Research and Development Center.
Dolkart, J. L. (1992, March), \\tafcing p^per on legal reform in the area of sexual harassmeat:
Procedural and evidentiary reform. In Northwest Women's Law Center, Sex and power in the
137
Lazanis, R. S,, & Folkman, S, (1984), Stress, apprtusal and coping. New York: Springer,
Livingston, J, A, (1982), Responses to sexual harassment on the job: Legal, organizational aod
individual actions, Journai of Social Issues, 38(4), 5-22,
Loy, P, H,, & Stewart, L, P, (1984), The extent and effects of sexual harassment of woiidng womoi.
Sociological
Focus,
17, 3]-43,
Maitindale, M, (1990), Sexual harassment in the miliiary: 1988. Arlington, VA: Defense Manpower
Data Center,
Maypole, D, (1986), Sexual harassment of social workers at work. Social Work, 31, 29-34.
McKinney, L., Olson, C, V,, & Satlerfield, A, (1988), Graduate students' experiences with and
responses to sexual harassment: A research note, Joumal of Interpersonal Violence, 3, 319-
325,
Meltebeke v. Labor and Industries, S52 P, 2d 859 (Or. App, CT, 1993).
Meritor Savings Bank v. VinsOD, 477 U.S, 57 (1986) (No. 84-1979),
Ormerod, A. J, (1991), The effect cf
self-efficacy
and outcome expectations on responses to sexual
harassment. Unpublished master's equivalency thesis. Department of Educational I^ycholo-
gy. University of niinms. Champaign,
Perlin, L,, & Schooler, C, (1978), The structure of coping. Joumal of Health and Social Behavior,
19, 2-21,
PhilUps, C, M,, Stockdale, J, E,, A Joeman, L, M. (1989). The risks qf going to work. London: The
Suzy Lamplugh Trust,
Pryor, J, B,, LaVite, C, M., & Stoller, L. M, (1993). A social psychological analysis of sexual
harassment: The person/situation interaction, Joumal of Vocational Behavior, 42, 68-83,
Rabidue v, Osceoia Refining Co,, 584 F, Supp, 419 (E. D. Mich, 1984),
Rabmowitz, V, C, (1990). Coping with sexual harassment. In M, A, Paludi (Ed,), Ivory power:
Sexual harassment on campus (pp. J03-118), Albany: SUNY Press.
Reed v, Shephard, 929 F. 2d 484 (7th Cir, 1991).
Robinson v. Jacksonville Shipyards,
IDC.
59 L Wl 2470 (1991) (DC M Fla).
Salisbury, J., Ginorio, A. B,, Remick, H., & Stringer, D, M, (1986), Counseling victims of sexual
harassment. Psychotherapy, 23, 316-324,
Saunders, B, E, (1992, Oct, 23), Sexual harassment of women in the workplace:
Results
from the
national women's
study.
Presentation
to the 8th Annual NC/SC Labor Law Seminar, AshvUle,
NC,
Schneider, B, E, (1991). Put up and ^ut
up:
Woikplace sexual assaults. Gender and
Society,
5, 533-
548.
Shaney, M. J. (1986), Pocqxions of harm: The consent defense in sexual harassment cases, Iowa
Law Review, 71, 1109-1135,
Terpstra, D. E., & Baker, D. D. (1988). Outcomes of sexual harassment charges. Academy qf
Management Journal, 31, 185-194,
Terpstra, D. E,, & Baker, D. D, (1989), The identification and classification of responses to sexual
harasstnent, Joumal of Organizational Behavior, 10, 1-14.
Thacker, R, A., & Ferris, G, R, (1991), Understanding sexual harassment in the workplace: Tbe
influence of power and politics within the dyadic interaction of harasser and target. Human
Resource Management
Review,
1, 23-37.
Till, F, J, (1980), Sexual harassment: A report on the sexual harassment cf students. Washington,
DC:
National Advisory Council on Women's Educational Program.
Ukarish v. Magnesium Elektron, 31 FairEmpl. Prac, Cas (BNA) 1315 (D.D.C, 1983),
V.S. V, Lanier, Western District oflfcnnessee, CR #92-2OI72-TU (1992),
Vance v. Southern BeU Tkl, &Tel, Company, 863 F.2d 1503 (llth Cir. 1989).
U.S. Merit Systems Protection Board. (1981). Sexualhamssmem/^federal workers:Is itaproblem?
Washington, IX:: U.S, Govemment Printing Office,
U.S, Merit SystenK PititecdoD Board (1S)88), Sexual harassment of federal workers: An updme.
Washington, DC: U,S, Govemment Printing Office,
Wagner, E. S, (1992), Sexual harassment in tke workplace: How to prevent, investigate and resolve
con^>tamts in your orgamzation. New Y«k: AMACOM,
r, S. (1983), Shifting die communication burden: A jneaniogfu) consent standard in rape.
Harvard
V/omm's
Law
Jawtial,
6, 143-170.
138
LOUISE F. FITZGERALD PH.D., is Associate Professor of PsydjoVogy and
Womens' Studies at the University of Illinois at Urbana—Champa^. She re-
ceived hei Ph.D. in psychology from the Ohio State University ia 1979 and has
held faculty positions at Kent State University and the University of Califomia—
Santa Baibara, Her major research area is sexual harassment in the workplace
and academia, which she has been studying for nearly a decade. She is the 1992
recipient of the Holland ftize for Excellence in Research in Personality and
Career Development and in 1991 was the psychological consultant to Professor
Anita Hill's legal team during the Clarence Thomas confirmation hearings.
SUZANNE SWAN, M.A., is a doctoral candidate in personality and social
ecology at the University of Illinois at Urbana—Champaign.
KARLA FISCHER, PH.D., J.D., is Assistant Professor of PSychoiogy: Social
and Health Sciences, and Assistant Professor of Law at Duke University, She
received both her Ph.D. in psychology and her J.D, from the University of
Illinois. Her major interests arc the psychological impact of participation in the
legal system and violence against women. She is currently focusing on a study of
court protection orders, for which she received SPSSI's 1993 Social Issues
Dissertation Award.
... A considerable body of research has explored the existence and effectiveness of policies and procedures of sexual harassment, generally comparing those policies with model policies (Reese & Lindenberg, 2002) or desirable features that make a policy effective. Research has also extensively explored the low formal reporting of sexual harassment incidents and the related reasons (Fitzgerald, Swan, & Fischer, 1995;Mani, 2004). There are also instances where research has explored how victims' rights are protected or how due process is followed in sexual harassment cases (Hart, 2012). ...